Spatialized audio is sound that is processed to give the listener an impression of a sound source within a three-dimensional environment. A more realistic experience is observed when listening to spatialized sound than stereo because stereo only varies across one axis, usually the x (horizontal) axis.
In the past, binaural sound from headphones was the most common approach to spatialization. The use of headphones takes advantage of the lack of crosstalk and a fixed position between sound source (the speaker driver) and the ear. Gradually, these factors are endowed upon conventional loudspeakers through more sophisticated digital signal processing. The wave of multimedia computer content and equipment has increased the use of stereo speakers in conjunction with microcomputers. Additionally, complex audio signal processing equipment, and the current consumer excitement surrounding the computer market, increases the awareness and desire for quality audio content. Two speakers, one on either side of a personal computer, carry the particular advantage of having the listener sitting rather closely and in an equidistant position between the speakers. The listener is probably also sitting down, therefore moving infrequently. This typical multimedia configuration probably comes as close to binaural sound using headphones as can be expected from free field speakers, increasing the probability of success for future spatialization systems.
Spatial audio can be useful whenever a listener is presented with multiple auditory streams. Spatial audio requires information about the positions of all events that need to be audible, including those outside of the field of vision, or that would benefit from increased immersion in an environment. Possible applications of spatial audio processing techniques include:                Military communication systems to and between individuals within military vehicles, ships and aircraft as well as to and between dismounted soldiers;        complex supervisory control systems such as telecommunications and air traffic control systems;        complex supervisory control system such as telecommunications and air traffic control systems;        civil and military aircraft warning systems;        teleconferencing and telepresence applications;        virtual and augmented reality environments;        computer-user interfaces and auditory displays, especially those intended for use by the visually impaired;        personal information and guidance systems such as those used to provide exhibit information to visitors in a museum;        arts and entertainment, especially video games and music, to name but a few.        
Environmental cues, such as early echoes and dense reverberation, are important for a realistic listening experience and are known to improve localization and externalization of audio sources. However, the cost of exact environmental modeling is extraordinarily high. Moreover, existing spatial audio systems are designed for use via headphones. This requirement may result in certain limitations on their use. For example, spatial audio may be limited to those applications for which a user is already wearing some sort of headgear, or for which the advantages of spatial sound outweigh the inconvenience of a headset.
U.S. Pat. Nos. 5,272,757, 5,459,790, 5,661,812, and 5,841,879, all to Scofield disclose head mounted surround sound systems. However, none of the Scofield systems appear to use head related transfer function (HRTF) filtering to produce spatialized audio signals. Furthermore, Scofield uses a system that converts signals from a multiple surround speaker system to a pair of signals for two speakers. This system appears to fail a real-time spatialization system where a person's head position varies in orientation and azimuth, thus requiring adjustment in filtering in order to maintain appropriate spatial locations.
One current method for generating spatialized audio is to use multiple speaker panning. This method only works for listeners positioned at a sweet spot within the speaker array. This method cannot be used for mobile applications. Another method, often used with headphones, requires complex individual filters or synthesized sound reflections. This method performs filtering of a monaural source with a pair of filters defined by a pair of head related transfer functions (HRTFs) for a particular location. Each of these methods have limitations and disadvantages. The latter method works best if individual filters are used, but the procedure to produce individual filters is complex. Further, if individual filters or synthesized sound reflections are not used, then front-back confusions and poor externalization of the sound source would result. Thus, there is a need to overcome the above-identified problems.
Accordingly, the present invention provides a solution to overcome the above problems. In the present invention, a pair of speakers is mounted in a location near the temple of a listener's head, such for example, on an eyeglass frame or inside a helmet, rather than in headphones. A head tracking system also mounted on the frame where speakers are mounted determines the location and orientation of the listener's head and provides the measurements to a computer system for audio signal processing in conjunction with a head related transfer function (HRTF) filter to produce spatialized audio. The HRTF filter maintains virtual location of the audio signals, thus allowing the listener to change locations and head orientation without degradation of the audio signal. The system of the present invention produces virtual sound sources that are externally perceived and positioned at any desired orientation in azimuth and elevation from the listener.
In its broader aspects, the present invention provides an apparatus for producing spatialized audio, the apparatus comprising at least one pair of speakers positioned near a user's temple for generating spatialized audio signals, whereby the speakers are positioned coaxially with a user's ear regardless of the user's head movement; a tracking system for tracking the user's head orientation and location; a head related transfer function (HRTF) filter for maintaining virtual location of the audio signals thereby allowing the user to change location and head orientation without degradation of the virtual location of audio signals; and a processor for receiving signals from the tracking system and causing the filter to generate spatialized audio, wherein the speakers are positioned to generate frontal positioning cues to augment spatial filtering for virtual frontal sources without degrading spatial filtering for other virtual positions.
In another aspect, a method of producing spatialized audio signals, the method comprising: positioning at least one pair of speakers near a user's temple for generating spatialized audio signals, whereby the speakers are positioned coaxially with a user's ear regardless of the user's head movement to generate frontal positioning cues to augment spatial filtering for virtual frontal sources without degrading spatial filtering for other virtual positions; tracking orientation and location of the user's head using a tracking system; maintaining virtual location of the audio signals using a head related transfer function (HRTF) filter; and processing signals received from the tracking system using a processor; and controlling the filter using the processor to generate spatialized audio signals.
In a further aspect, the present invention provides a system for producing spatialized audio signals, the system comprising: means for positioning at least one pair of speakers near a user's temple for generating spatialized audio signals, whereby the speakers are positioned coaxially with a user's ear regardless of the user's head movement; a tracking means for tracking orientation and location of the user's head; a filtering means for maintaining virtual location of the audio signals; and means for processing signals received from the tracking means; and means for controlling the filter means to generate spatialized audio signals.